Instructions to use Unbabel/Tower-Plus-72B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Unbabel/Tower-Plus-72B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Unbabel/Tower-Plus-72B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Unbabel/Tower-Plus-72B")
model = AutoModelForCausalLM.from_pretrained("Unbabel/Tower-Plus-72B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Local Apps Settings

vLLM

How to use Unbabel/Tower-Plus-72B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Unbabel/Tower-Plus-72B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Unbabel/Tower-Plus-72B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Unbabel/Tower-Plus-72B

SGLang

How to use Unbabel/Tower-Plus-72B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Unbabel/Tower-Plus-72B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Unbabel/Tower-Plus-72B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Unbabel/Tower-Plus-72B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Unbabel/Tower-Plus-72B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Unbabel/Tower-Plus-72B with Docker Model Runner:
```
docker model run hf.co/Unbabel/Tower-Plus-72B
```

Improve model card: Add pipeline tag, paper and project page links

by nielsr HF Staff - opened Jul 2, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+13

-8

Files changed (1) hide show

README.md +13 -8

README.md CHANGED Viewed

@@ -1,6 +1,5 @@
 ---
 base_model: Qwen/Qwen2.5-72B
-license: cc-by-nc-sa-4.0
 language:
 - de
 - nl
@@ -25,29 +24,35 @@ language:
 - ro
 - fi
 library_name: transformers
 ---
 ![Tower Plus Pareto](./Tower-plus-pareto.png)
 # Model Description:
 **Tower+ 72B** is build on top of Qwen 2.5 72B. The model goes through the Continuous Pretraining (CPT), Instruction Tuning (IT) and Weighted Preference Optimization (WPO). During all these stages we include parallel and multilingual data (covering 22 languages).
-- **Developed by:** Unbabel
-- **Model type:** A 72B parameter model fine-tuned on a mix of _translation-related tasks_ as well as  _general instruction-following_ datasets that include reasoning, code instructions, etc.
-- **Languages:** German, Spanish, French, Italian, Korean, Dutch, Russian, English, Portuguese (Portugal), Portuguese (Brazilian), Spanish (Latin America), Chinese (Simplified), Chinese (Traditional), Czech, Ukrainian, Hindi, Icelandic, Japanese, Polish, Swedish, Hungarian, Romanian, Danish, Norwegian (Nynorsk), Norwegian (Bokmål), Finnish
-- **License:** CC-BY-NC-4.0
-- **Context Size:**: 131,072 tokens (recommended generation tokens 8192)
 # Intended uses & limitations
-Tower is intended for multilingual tasks and its specially strong on translation related tasks.
 Another usecase Tower works well is for creating multilingual synthethic data (for the languages it covers). You can do this either by translating instructions and the respective answers or by asking the model to create an instruction given a document as seed data.
 # Usage:
-When using the model, make sure your prompt is formated correctly!
 Also, we recommend using VLLM rather than Hugging Face.

 ---
 base_model: Qwen/Qwen2.5-72B
 language:
 - de
 - nl
 - ro
 - fi
 library_name: transformers
+license: cc-by-nc-sa-4.0
+pipeline_tag: text-generation
 ---
+This repository contains the Tower+ 72B model, as presented in the paper [Tower+: Bridging Generality and Translation Specialization in Multilingual LLMs](https://huggingface.co/papers/2506.17080).
+Project Page: [https://huggingface.co/collections/Unbabel/tower-plus-6846ca452a10c0905dc03c0f](https://huggingface.co/collections/Unbabel/tower-plus-6846ca452a10c0905dc03c0f)
 ![Tower Plus Pareto](./Tower-plus-pareto.png)
 # Model Description:
 **Tower+ 72B** is build on top of Qwen 2.5 72B. The model goes through the Continuous Pretraining (CPT), Instruction Tuning (IT) and Weighted Preference Optimization (WPO). During all these stages we include parallel and multilingual data (covering 22 languages).
+-   **Developed by:** Unbabel
+-   **Model type:** A 72B parameter model fine-tuned on a mix of _translation-related tasks_ as well as _general instruction-following_ datasets that include reasoning, code instructions, etc.
+-   **Languages:** German, Spanish, French, Italian, Korean, Dutch, Russian, English, Portuguese (Portugal), Portuguese (Brazilian), Spanish (Latin America), Chinese (Simplified), Chinese (Traditional), Czech, Ukrainian, Hindi, Icelandic, Japanese, Polish, Swedish, Hungarian, Romanian, Danish, Norwegian (Nynorsk), Norwegian (Bokmål), Finnish
+-   **License:** CC-BY-NC-4.0
+-   **Context Size:**: 131,072 tokens (recommended generation tokens 8192)
 # Intended uses & limitations
+Tower is intended for multilingual tasks and its specially strong on translation related tasks.
 Another usecase Tower works well is for creating multilingual synthethic data (for the languages it covers). You can do this either by translating instructions and the respective answers or by asking the model to create an instruction given a document as seed data.
 # Usage:
+When using the model, make sure your prompt is formated correctly!
 Also, we recommend using VLLM rather than Hugging Face.